Anthropic released Cowork on Monday, a brand new AI agent capability that extends the facility of its wildly successful Claude Code tool to non-technical users — and in keeping with company insiders, the team built your complete feature in roughly per week and a half, largely using Claude Code itself.
The launch marks a significant inflection point within the race to deliver practical AI agents to mainstream users, positioning Anthropic to compete not only with OpenAI and Google in conversational AI, but with Microsoft's Copilot within the burgeoning marketplace for AI-powered productivity tools.
"Cowork helps you to complete non-technical tasks very similar to how developers use Claude Code," the company announced via its official Claude account on X. The feature arrives as a research preview available exclusively to Claude Max subscribers — Anthropic's power-user tier priced between $100 and $200 per 30 days — through the macOS desktop application.
For the past 12 months, the industry narrative has focused on large language models that may write poetry or debug code. With Cowork, Anthropic is betting that the actual enterprise value lies in an AI that may open a folder, read a messy pile of receipts, and generate a structured expense report without human hand-holding.
How developers using a coding tool for vacation research inspired Anthropic's latest product
The genesis of Cowork lies in Anthropic's recent success with the developer community. In late 2024, the corporate released Claude Code, a terminal-based tool that allowed software engineers to automate rote programming tasks. The tool was a success, but Anthropic noticed a peculiar trend: users were forcing the coding tool to perform non-coding labor.
In keeping with Boris Cherny, an engineer at Anthropic, the corporate observed users deploying the developer tool for an unexpectedly diverse array of tasks.
"Since we launched Claude Code, we saw people using it for all styles of non-coding work: doing vacation research, constructing slide decks, cleansing up your email, cancelling subscriptions, recovering wedding photos from a harddrive, monitoring plant growth, controlling your oven," Cherny wrote on X. "These use cases are diverse and surprising — the rationale is that the underlying Claude Agent is one of the best agent, and Opus 4.5 is one of the best model."
Recognizing this shadow usage, Anthropic effectively stripped the command-line complexity from their developer tool to create a consumer-friendly interface. In its blog post announcing the feature, Anthropic explained that developers "quickly began using it for nearly every thing else," which "prompted us to construct Cowork: a less complicated way for anyone — not only developers — to work with Claude in the exact same way."
Contained in the folder-based architecture that lets Claude read, edit, and create files in your computer
Unlike a normal chat interface where a user pastes text for evaluation, Cowork requires a distinct level of trust and access. Users designate a selected folder on their local machine that Claude can access. Inside that sandbox, the AI agent can read existing files, modify them, or create entirely latest ones.
Anthropic offers several illustrative examples: reorganizing a cluttered downloads folder by sorting and intelligently renaming each file, generating a spreadsheet of expenses from a group of receipt screenshots, or drafting a report from scattered notes across multiple documents.
"In Cowork, you give Claude access to a folder in your computer. Claude can then read, edit, or create files in that folder," the corporate explained on X. "Try it to create a spreadsheet from a pile of screenshots, or produce a primary draft from scattered notes."
The architecture relies on what’s referred to as an "agentic loop." When a user assigns a task, the AI doesn’t merely generate a text response. As an alternative, it formulates a plan, executes steps in parallel, checks its own work, and asks for clarification if it hits a roadblock. Users can queue multiple tasks and let Claude process them concurrently — a workflow Anthropic describes as feeling "much less like a back-and-forth and rather more like leaving messages for a coworker."
The system is built on Anthropic's Claude Agent SDK, meaning it shares the identical underlying architecture as Claude Code. Anthropic notes that Cowork "can tackle lots of the same tasks that Claude Code can handle, but in a more approachable form for non-coding tasks."
The recursive loop where AI builds AI: Claude Code reportedly wrote much of Claude Cowork
Perhaps probably the most remarkable detail surrounding Cowork's launch is the speed at which the tool was reportedly built — highlighting a recursive feedback loop where AI tools are getting used to construct higher AI tools.
During a livestream hosted by Dan Shipper, Felix Rieseberg, an Anthropic worker, confirmed that the team built Cowork in roughly per week and a half.
Alex Volkov, who covers AI developments, expressed surprise on the timeline: "Holy shit Anthropic built 'Cowork' within the last… week and a half?!"
This prompted immediate speculation about how much of Cowork was itself built by Claude Code. Simon Smith, EVP of Generative AI at Klick Health, put it bluntly on X: "Claude Code wrote all of Claude Cowork. Can all of us agree that we're in no less than somewhat of a recursive improvement loop here?"
The implication is profound: Anthropic's AI coding agent could have substantially contributed to constructing its own non-technical sibling product. If true, that is some of the visible examples yet of AI systems getting used to speed up their very own development and expansion — a technique that would widen the gap between AI labs that successfully deploy their very own agents internally and people who don’t.
Connectors, browser automation, and skills extend Cowork's reach beyond the local file system
Cowork doesn't operate in isolation. The feature integrates with Anthropic's existing ecosystem of connectors — tools that link Claude to external information sources and services comparable to Asana, Notion, PayPal, and other supported partners. Users who’ve configured these connections in the usual Claude interface can leverage them inside Cowork sessions.
Moreover, Cowork can pair with Claude in Chrome, Anthropic's browser extension, to execute tasks requiring web access. This mix allows the agent to navigate web sites, click buttons, fill forms, and extract information from the web — all while operating from the desktop application.
"Cowork includes quite a lot of novel UX and safety features that we predict make the product really special," Cherny explained, highlighting "a built-in VM [virtual machine] for isolation, out of the box support for browser automation, support for all of your claude.ai data connectors, asking you for clarification when it's unsure."
Anthropic has also introduced an initial set of "skills" specifically designed for Cowork that enhance Claude's ability to create documents, presentations, and other files. These construct on the Skills for Claude framework the corporate announced in October, which provides specialized instruction sets Claude can load for particular sorts of tasks.
Why Anthropic is warning users that its own AI agent could delete their files
The transition from a chatbot that implies edits to an agent that makes edits introduces significant risk. An AI that may organize files can, theoretically, delete them.
In a notable display of transparency, Anthropic devoted considerable space in its announcement to warning users about Cowork's potential dangers — an unusual approach for a product launch.
The corporate explicitly acknowledges that Claude "can take potentially destructive actions (comparable to deleting local files) if it's instructed to." Because Claude might occasionally misinterpret instructions, Anthropic urges users to offer "very clear guidance" about sensitive operations.
More concerning is the chance of prompt injection attacks — a way where malicious actors embed hidden instructions in content Claude might encounter online, potentially causing the agent to bypass safeguards or take harmful actions.
"We've built sophisticated defenses against prompt injections," Anthropic wrote, "but agent safety — that’s, the duty of securing Claude's real-world actions — continues to be an energetic area of development within the industry."
The corporate characterised these risks as inherent to the present state of AI agent technology reasonably than unique to Cowork. "These risks aren't latest with Cowork, but it surely is perhaps the primary time you're using a more advanced tool that moves beyond an easy conversation," the announcement notes.
Anthropic's desktop agent strategy sets up a direct challenge to Microsoft Copilot
The launch of Cowork places Anthropic in direct competition with Microsoft, which has spent years attempting to integrate its Copilot AI into the material of the Windows operating system with mixed adoption results.
Nonetheless, Anthropic's approach differs in its isolation. By confining the agent to specific folders and requiring explicit connectors, they try to strike a balance between the utility of an OS-level agent and the safety of a sandboxed application.
What distinguishes Anthropic's approach is its bottom-up evolution. Moderately than designing an AI assistant and retrofitting agent capabilities, Anthropic built a robust coding agent first — Claude Code — and is now abstracting its capabilities for broader audiences. This technical lineage may give Cowork more robust agentic behavior from the beginning.
Claude Code has generated significant enthusiasm amongst developers since its initial launch as a command-line tool in late 2024. The corporate expanded access with a web interface in October 2025, followed by a Slack integration in December. Cowork is the subsequent logical step: bringing the identical agentic architecture to users who may never touch a terminal.
Who can access Cowork now, and what's coming next for Windows and other platforms
For now, Cowork stays exclusive to Claude Max subscribers using the macOS desktop application. Users on other subscription tiers — Free, Pro, Team, or Enterprise — can join a waitlist for future access.
Anthropic has signaled clear intentions to expand the feature's reach. The blog post explicitly mentions plans so as to add cross-device sync and convey Cowork to Windows as the corporate learns from the research preview.
Cherny set expectations appropriately, describing the product as "early and raw, much like what Claude Code felt like when it first launched."
To access Cowork, Max subscribers can download or update the Claude macOS app and click on on "Cowork" within the sidebar.
The true query facing enterprise AI adoption
For technical decision-makers, the implications of Cowork extend beyond any single product launch. The bottleneck for AI adoption is shifting — not is model intelligence the limiting factor, but reasonably workflow integration and user trust.
Anthropic's goal, as the corporate puts it, is to make working with Claude feel less like operating a tool and more like delegating to a colleague. Whether mainstream users are able to hand over folder access to an AI that may misinterpret their instructions stays an open query.
However the speed of Cowork's development — a significant feature in-built ten days, possibly by the corporate's own AI — previews a future where the capabilities of those systems compound faster than organizations can evaluate them.
The chatbot has learned to make use of a file manager. What it learns to make use of next is anyone's guess.
